Microsoft’s multi-agent AI system tops Anthropic’s Mythos on cybersecurity benchmark

vm_admin

17 hours ago

Microsoft’s multi-agent AI system tops Anthropic’s Mythos on cybersecurity benchmark

By Todd Bishop
Publication Date: 2026-05-14 00:16:00

by Todd Bishop on May 13, 2026 at 5:16 pmMay 14, 2026 at 9:05 am

CyberGym benchmark scores over time, showing the rapid improvement in AI vulnerability discovery capabilities. Microsoft’s multi-model MDASH system (top right) tops the leaderboard at 88.4%. (CyberGym / UC Berkeley)

Mythos has been MDASH’d.

A new AI-powered system from Microsoft surpassed a headline-grabbing rival from Anthropic on a leading cybersecurity benchmark, using more than 100 specialized AI agents working together across multiple AI models to find real-world software vulnerabilities.

Microsoft’s system, codenamed MDASH, was introduced this week alongside the disclosure of 16 new vulnerabilities it found in different versions of Windows, including four “critical” remote code execution flaws fixed in this month’s Patch Tuesday release.

The company, which has faced persistent criticism…